feat: allow gz unfiltered permit list via niffler

Cargo.toml

@@ -87,6 +87,7 @@ clap = { version = "4.5.23", features = [
 noodles = { version = "0.85.0", features = ["bam", "bgzf", "sam"] }
 dashmap = { version = "6.1.0", features = ["serde", "inline"] }
 nalgebra = "0.33.2"
+niffler = { version = "2.6.0", default-features = false, features = ["gz", "gz_cloudflare_zlib"] }
 
 [profile.release]
 #debug = true

scripts/testing/compare_counts.py

@@ -55,9 +55,9 @@ def compare_quants(args):
 
 		odict = { "nobs_ref" : a.n_obs, "nobs_test" : b.n_obs }
 
-		odict['diff_U'] = float(abs(a.layers['unspliced'] - b[ a.obs_names, : ].layers['unspliced']).sum())
-		odict['diff_S'] = float(abs(a.layers['spliced'] - b[ a.obs_names, : ].layers['spliced']).sum())
-		odict['diff_A'] = float(abs(a.layers['ambiguous'] - b[ a.obs_names, : ].layers['ambiguous']).sum())
+		odict['diff_U'] = float(abs(a.layers['unspliced'] - b[ a.obs_names, a.var_names ].layers['unspliced']).sum())
+		odict['diff_S'] = float(abs(a.layers['spliced'] - b[ a.obs_names, a.var_names ].layers['spliced']).sum())
+		odict['diff_A'] = float(abs(a.layers['ambiguous'] - b[ a.obs_names, a.var_names ].layers['ambiguous']).sum())
 
 		odict['obs_ref-obs_test'] = list(a.obs_names.difference(b.obs_names))
 		odict['obs_test-obs_ref'] = list(b.obs_names.difference(a.obs_names))
@@ -67,7 +67,7 @@ def compare_quants(args):
 
 		odict = { "nobs_ref" : a.n_obs, "nobs_test" : b.n_obs }
 
-		odict['diff_X'] = float(abs(a.X - b[ a.obs_names, : ].X).sum())
+		odict['diff_X'] = float(abs(a.X - b[ a.obs_names, a.var_names ].X).sum())
 
 		odict['obs_ref-obs_test'] = list(a.obs_names.difference(b.obs_names))
 		odict['obs_test-obs_ref'] = list(b.obs_names.difference(a.obs_names))

src/atac/cellfilter.rs

@@ -397,8 +397,17 @@ pub fn generate_permit_list(gpl_opts: GenPermitListOpts) -> anyhow::Result<u64>
     }
 
     if let CellFilterMethod::UnfilteredExternalList(fname, _) = &filter_meth {
-        let i_file = File::open(fname).context("could not open input file")?;
-        let br = BufReader::new(i_file);
+        let (reader, compression) = niffler::from_path(fname)
+            .with_context(|| format!("coult not open input file {}", fname.display()))?;
+        let br = BufReader::new(reader);
+
+        info!(
+            log,
+            "reading permit list from {}; inferred format {:#?}",
+            fname.display(),
+            compression
+        );
+
         unfiltered_bc_counts = Some(populate_unfiltered_barcode_map(br, &mut first_bclen, rc));
         info!(
             log,

src/cellfilter.rs

@@ -26,6 +26,7 @@ use libradicl::rad_types::{self, RadType};
 use libradicl::BarcodeLookupMap;
 use libradicl::{chunk, record::AlevinFryReadRecord};
 use needletail::bitkmer::*;
+use niffler::{compression, Error};
 use num_format::{Locale, ToFormattedString};
 use serde::Serialize;
 use serde_json::json;
@@ -634,8 +635,17 @@ pub fn generate_permit_list(gpl_opts: GenPermitListOpts) -> anyhow::Result<u64>
     let mut first_bclen = 0usize;
     let mut unfiltered_bc_counts = None;
     if let CellFilterMethod::UnfilteredExternalList(fname, _) = &filter_meth {
-        let i_file = File::open(fname).context("could not open input file")?;
-        let br = BufReader::new(i_file);
+        let (reader, compression) = niffler::from_path(fname)
+            .with_context(|| format!("coult not open input file {}", fname.display()))?;
+        let br = BufReader::new(reader);
+
+        info!(
+            log,
+            "reading permit list from {}; inferred format {:#?}",
+            fname.display(),
+            compression
+        );
+
         unfiltered_bc_counts = Some(populate_unfiltered_barcode_map(br, &mut first_bclen));
         info!(
             log,